© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67321
    Prev
    Next
    INDEX
    Explanations

    physical locations or environments

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects words denoting physical settings, environments, locations, or compartments.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Bewohner
    -0.93
    lehrer
    -0.92
     ویکی
    -0.91
     대한
    -0.90
     wines
    -0.90
    Tienda
    -0.90
    lias
    -0.89
    cising
    -0.89
     parry
    -0.89
    Nadie
    -0.88
    POSITIVE LOGITS
     or
    2.06
     where
    1.45
     without
    1.02
     other
    0.96
    even
    0.94
    other
    0.90
     will
    0.90
     hvor
    0.88
    filer
    0.87
    专用
    0.87
    Activations Density 0.125%

    No Known Activations